JP6816164B2 - Technology for behavioral pairing in contact center systems - Google Patents

Description

(Refer to related applications)
This international patent application claims priority over U.S. Patent Application No. 15 / 582,223 filed on April 28, 2017, and U.S. Patent Application No. 15 filed on April 28, 2017. Claiming priority over U.S. Patent Application No. 15 / 691,106 filed on August 30, 2017, which is a continuation of / 582,223, each U.S. patent application is fully described herein. The whole is incorporated herein by reference in its entirety.

(Field of disclosure)
The present disclosure relates generally to pairing contacts and agents in a contact center, and more specifically to behavioral pairing techniques in a contact center system.

(Background of disclosure)
A typical contact center algorithmically assigns contacts arriving at the contact center to agents available to process those contacts. At some point, a contact center may have agents that are available and waiting to be assigned to inbound or outbound contacts (eg, calls, internet chat sessions, email). At another time, the contact center may have contacts waiting in one or more queues for agents to become available for assignment.

In some typical contact centers, contacts are assigned to agents ordered based on incoming time, and agents receive time-ordered contacts when those agents become available. To do. This strategy can be referred to as a "first in, first out", "FIFO", or "round robin" strategy. In other typical contact centers, other measures such as "performance-based routing" or "PBR" measures may be used.

In other more progressive contact centers, contacts are paired with agents by using "behavioral pairing" or "BP" measures, and under "behavioral pairing" or "BP" measures. The contacts and agents follow, so that when the profits of all allocations under the BP strategy are summed, the profits outweigh the profits of other measures such as FIFA and performance-based routing (“PBR”) measures. Can be paired carefully (priority) in a manner that allows the assignment of contact-agent pairs. While the BP is designed to promote balanced utilization (or degree of utilization distortion) of agents in the skill queue, it nevertheless goes beyond what a FIFO or PBR method would allow. Simultaneously improve overall contact center performance. This is a great achievement as the BP acts on the same calls and agents as the FIFO or PBR method and utilizes the agents approximately equally as the FIFO provides, while improving overall contact center performance. Is. BP is described, for example, in US Pat. No. 9,300,802, incorporated herein by reference. Additional information about these features and other features regarding pairing or matching modules (sometimes also referred to as "SATMAP", "routing system", "routing engine", etc.) can be found, for example, by reference herein. It is described in the incorporated US Pat. No. 8,879,715.

The BP strategy may use one-dimensional ordering of contact types and agents, along with diagonal strategies to determine the preferred pairing. However, this strategy can limit, or otherwise limit, the type and number of variables that the BP policy can optimize, or the amount that one or more variables can be optimized, given greater degrees of freedom. ..

In view of the above, it can be understood that there is a need for a system that allows to improve the efficiency and performance of pairing strategies designed to choose from multiple possible pairings, such as BP strategies.

U.S. Pat. No. 9,300,802 U.S. Pat. No. 8,879,715

(Summary of disclosure)
Techniques for behavioral pairing in contact center systems are disclosed. In one particular embodiment, the technique can be realized as a method for behavioral pairing in a contact center system, the method being communicatively coupled to the contact center system and configured to operate in the contact center system. At least one computer processor is used to determine the multiple agents available for a connection to a contact, and at least one computer processor is capable of multiple pairings between a contact and multiple agents. Determining preferred contact-agent pairing and selecting one of a plurality of preferred contact-agent pairings according to a probabilistic model by at least one computer processor and by at least one computer processor. Includes outputting a selected one of a plurality of preferred contact-agent pairings for connectivity in a contact center system.

According to other aspects of this particular embodiment, the probabilistic model is a network flow model for balancing agent utilization, a network flow model for applying the amount of agent utilization distortion, and the overall of at least one contact center metric. It can be a network flow model for optimizing expected values. Also, at least one contact center metric can be at least one of revenue generation, customer satisfaction, and average processing time.

According to other aspects of this particular embodiment, the probabilistic model can be a network flow model constrained by agent skill and contact skill needs. In addition, the network flow model can be adjusted to minimize agent utilization imbalances according to the constraints of agent and contact skill needs.

According to other aspects of this particular embodiment, the probabilistic model may incorporate expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.

In another particular embodiment, the technique can be implemented as a system for behavioral pairing in a contact center system, the system comprising at least one computer processor configured to operate in the contact center system. At least one computer processor is configured to perform the steps in the above method.

In another particular embodiment, the technique can be implemented as a product for behavioral pairing in a contact center system, the product being a non-temporary processor readable medium and instructions stored on the medium. The instructions are configured to be readable from the medium by at least one computer processor configured to operate in a contact center system, thereby at least performing the steps in the above method. Run one computer processor.

The present disclosure will be described in more detail with reference to specific embodiments as shown in the accompanying drawings. Although the present disclosure is described below with reference to specific embodiments, it should be understood that the present disclosure is not limited to that particular embodiment. Additional implementations, modifications and embodiments, as well as other areas of use, that are within the scope of this disclosure as described herein and that this disclosure may be of significant use in this regard are described herein. Those skilled in the art who have access to the teachings in the book will recognize.
The present invention provides, for example,:
(Item 1)
A method for behavioral pairing in a contact center system
Determining multiple agents available for connection to a contact by at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system.
Determining a plurality of preferred contact-agent pairings from possible pairings between the contact and the plurality of agents by the at least one computer processor.
Selecting one of the plurality of preferred contact-agent pairings according to a probabilistic model by the at least one computer processor.
The at least one computer processor outputs the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
Including methods.
(Item 2)
The method according to item 1, wherein the probability model is a network flow model for balancing agent usage.
(Item 3)
The method according to item 1, wherein the probability model is a network flow model for applying the amount of agent utilization distortion.
(Item 4)
The method according to item 1, wherein the probabilistic model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 5)
The method of item 4, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 6)
The method according to item 1, wherein the probabilistic model is a network flow model constrained by agent skill and contact skill needs.
(Item 7)
6. The method of item 6, wherein the network flow model is adjusted to minimize agent utilization imbalances in accordance with the constraints of the agent skills and contact skill needs.
(Item 8)
The method of item 1, wherein the probabilistic model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.
(Item 9)
A system for behavioral pairing in a contact center system.
It comprises at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system, the at least one computer processor.
Determining multiple agents available for connecting to a contact,
Determining a plurality of preferred contact-agent pairings from possible pairings between the contact and the plurality of agents.
Choosing one of the plurality of preferred contact-agent pairings according to a probabilistic model,
To output the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
A system that is further configured to do.
(Item 10)
The system according to item 9, wherein the probability model is a network flow model for balancing the use of agents.
(Item 11)
The system according to item 9, wherein the probability model is a network flow model for applying the amount of agent utilization distortion.
(Item 12)
9. The system of item 9, wherein the probabilistic model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 13)
The system of item 12, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 14)
The system according to item 9, wherein the probability model is a network flow model constrained by agent skill and contact skill needs.
(Item 15)
The system of item 14, wherein the network flow model is tuned to minimize agent utilization imbalances according to the constraints of the agent skills and contact skill needs.
(Item 16)
9. The system of item 9, wherein the probabilistic model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.
(Item 17)
A product for behavioral pairing in a contact center system, the product of which is
Non-temporary processor readable media and
Instructions stored on the medium and
The instructions are configured to be readable from the medium by at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system. By
Determining multiple agents available for connecting to a contact,
Determining a plurality of preferred contact-agent pairings from possible pairings between the contact and the plurality of agents.
Choosing one of the plurality of preferred contact-agent pairings according to a probabilistic model,
To output the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
A product that operates the at least one computer processor to do so.
(Item 18)
The manufactured product according to item 17, wherein the probability model is a network flow model for balancing the use of agents.
(Item 19)
The manufactured product according to item 17, wherein the probability model is a network flow model for applying the amount of agent utilization distortion.
(Item 20)
The product according to item 17, wherein the probabilistic model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 21)
The product of item 20, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 22)
The manufactured product according to item 17, wherein the probabilistic model is a network flow model constrained by agent skill and contact skill needs.
(Item 23)
22. The product of item 22, wherein the network flow model is adjusted to minimize agent utilization imbalances according to the constraints of the agent skills and contact skill needs.
(Item 24)
The product of item 17, wherein the probabilistic model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.
(Item 25)
A method for behavioral pairing in a contact center system
Determining the multiple contacts available for a connection to an agent by at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system.
Determining a plurality of preferred contact-agent pairings from possible pairings between the agent and the plurality of contacts by the at least one computer processor.
Selecting one of the plurality of preferred contact-agent pairings according to a stochastic network flow model by the at least one computer processor.
The at least one computer processor outputs the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
Including methods.
(Item 26)
The method according to item 25, wherein the stochastic network flow model is a network flow model for balancing agent usage.
(Item 27)
The method according to item 25, wherein the probabilistic network flow model is a network flow model for applying the amount of agent utilization distortion.
(Item 28)
25. The method of item 25, wherein the probabilistic network flow model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 29)
28. The method of item 28, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 30)
25. The method of item 25, wherein the probabilistic network flow model is a network flow model constrained by agent skill and contact skill needs.
(Item 31)
30. The method of item 30, wherein the probabilistic network flow model is adjusted to minimize agent utilization imbalances according to the constraints of the agent skills and contact skill needs.
(Item 32)
25. The method of item 25, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.
(Item 33)
A system for behavioral pairing in a contact center system.
It comprises at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system, the at least one computer processor.
Determining the multiple contacts available for connecting to an agent,
Determining a plurality of preferred contact-agent pairings from possible pairings between the agent and the plurality of contacts.
Choosing one of the plurality of preferred contact-agent pairings according to a probabilistic network flow model,
To output the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
A system that is further configured to do.
(Item 34)
The system according to item 33, wherein the stochastic network flow model is a network flow model for balancing agent usage.
(Item 35)
The system according to item 33, wherein the stochastic network flow model is a network flow model for applying the amount of agent utilization distortion.
(Item 36)
33. The system of item 33, wherein the probabilistic network flow model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 37)
36. The system of item 36, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 38)
33. The system of item 33, wherein the probabilistic network flow model is a network flow model constrained by agent skill and contact skill needs.
(Item 39)
38. The system of item 38, wherein the probabilistic network flow model is tuned to minimize agent utilization imbalances according to the constraints of the agent skills and contact skill needs.
(Item 40)
33. The system of item 33, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.
(Item 41)
A product for behavioral pairing in a contact center system, the product of which is
Non-temporary processor readable media and
Instructions stored on the medium and
The instructions are configured to be readable from the medium by at least one computer processor communicatively coupled to the contact center system and configured to operate in the contact center system. By
Determining the multiple contacts available for connecting to an agent,
Determining a plurality of preferred contact-agent pairings from possible pairings between the agent and the plurality of contacts.
Choosing one of the plurality of preferred contact-agent pairings according to a probabilistic network flow model,
To output the selected one of the plurality of preferred contact-agent pairings for connection in the contact center system.
A product that operates the at least one computer processor to do so.
(Item 42)
The manufactured product according to item 41, wherein the stochastic network flow model is a network flow model for balancing the use of agents.
(Item 43)
The manufactured product according to item 41, wherein the stochastic network flow model is a network flow model for applying the amount of agent utilization distortion.
(Item 44)
The product according to item 41, wherein the probabilistic network flow model is a network flow model for optimizing the overall expected value of at least one contact center metric.
(Item 45)
The product of item 44, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time.
(Item 46)
The manufactured product according to item 41, wherein the stochastic network flow model is a network flow model constrained by agent skill and contact skill needs.
(Item 47)
46. The product of item 46, wherein the probabilistic network flow model is adjusted to minimize agent utilization imbalances in accordance with the constraints of the agent skills and contact skill needs.
(Item 48)
41. The product of item 41, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.

To facilitate a more complete understanding of the present disclosure, reference is made here to the accompanying drawings in which the same elements are referenced using the same numbers. These drawings should not be construed as limiting the disclosure and are intended to be merely exemplary.
FIG. 1 shows a block diagram of a contact center according to an embodiment of the present disclosure. FIG. 2 shows an example of a BP payout matrix according to an embodiment of the present disclosure. FIG. 3 represents an example of a simple BP utilization matrix according to an embodiment of the present disclosure. FIG. 4A shows an example of a BP skill-based payout matrix according to an embodiment of the present disclosure. FIG. 4B shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 4C shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 4D shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 4E shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 4F shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 4G shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5A represents an example of a BP skill-based payout matrix according to an embodiment of the present disclosure. FIG. 5B shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5C shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5D shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5E shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5F shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5G shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5H shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 5I shows an example of a BP network flow according to an embodiment of the present disclosure. FIG. 6 represents a flow diagram of a BP skill-based payout matrix method according to an embodiment of the present disclosure. FIG. 7A shows a flow diagram of a BP network flow method according to an embodiment of the present disclosure. FIG. 7B shows a flow diagram of a BP network flow method according to an embodiment of the present disclosure. FIG. 8 shows a flow chart of a BP network flow method according to an embodiment of the present disclosure. FIG. 9 shows a flow chart of a BP network flow method according to an embodiment of the present disclosure.

(Detailed explanation)
A typical contact center algorithmically assigns contacts arriving at the contact center to agents available to process those contacts. At some point, a contact center may have agents that are available and waiting to be assigned to inbound or outbound contacts (eg, calls, internet chat sessions, email). At another time, the contact center may have contacts waiting in one or more queues for agents to become available for assignment.

In some typical contact centers, contacts are assigned to agents ordered based on incoming time, and agents receive time-ordered contacts when those agents become available. To do. This strategy can be referred to as a "first in, first out", "FIFO", or "round robin" strategy. In other typical contact centers, other measures such as "performance-based routing" or "PBR" measures may be used.

In other more progressive contact centers, contacts are paired with agents by using "behavioral pairing" or "BP" measures, and under "behavioral pairing" or "BP" measures. The contacts and agents follow, so that when the profits of all allocations under the BP strategy are summed, the profits outweigh the profits of other measures such as FIFA and performance-based routing (“PBR”) measures. Can be paired carefully (priority) in a manner that allows the assignment of contact-agent pairs. While the BP is designed to promote balanced utilization (or degree of utilization distortion) of agents in the skill queue, it nevertheless goes beyond what a FIFO or PBR method would allow. Simultaneously improve overall contact center performance. This is a great achievement as the BP acts on the same calls and agents as the FIFO or PBR method and utilizes the agents approximately equally as the FIFO provides, while improving overall contact center performance. Is. BP is described, for example, in US Pat. No. 9,300,802, incorporated herein by reference. Additional information about these features and other features regarding pairing or matching modules (sometimes also referred to as "SATMAP", "routing system", "routing engine", etc.) can be found, for example, by reference herein. It is described in the incorporated US Pat. No. 8,879,715.

The BP strategy may use one-dimensional ordering of contact types and agents, along with diagonal strategies to determine the preferred pairing. However, this strategy can limit, or otherwise limit, the type and number of variables that the BP policy can optimize, or the amount that one or more variables can be optimized, given greater degrees of freedom. ..

In view of the above, it can be understood that there is a need for a system that allows to improve the efficiency and performance of pairing strategies designed to choose from multiple possible pairings, such as BP strategies. Such systems are optimized based on comparative advantage in execution time in some embodiments; maintaining uniform or approximately uniform agent utilization; to a single coherent model or a smaller number of coherent models. Aggregation of models across skills; can bring innumerable benefits, including the creation of more complex, high-performance, and competent models. As described in detail below, the technique can be multidimensional in nature (eg, multivariate) to determine linear programming, quadratic programming, or preferred contact-agent pairing. Other optimization techniques may be used. Examples of these techniques are, for example, 708-68 and 843-897 (Chapter 26, "Maximum Flow" and Chapter 29, "Linear Programming") (2009), third editions of Reference to Algonomics by Cormen et al. Described in 448-96 (2006) of "Technical Optimization" by Nocedal and Writ, the above references are incorporated herein by reference.

FIG. 1 shows a block diagram of a contact center 100 according to an embodiment of the present disclosure. The description herein describes how to simulate a contact center system that may include one or more modules and the network elements, computers, and / or components of the system. As used herein, the term "module" may be understood to refer to computing software, firmware, hardware, and / or various combinations thereof. However, a module should not be interpreted as software that is not implemented on hardware, firmware, or software that is not recorded on a processor-readable recordable storage medium (ie, the module itself is not software). It should be noted that the module is exemplary. Modules can be combined, integrated, separated and / or duplicated to support a variety of applications. Further, the functions described herein as performed in a particular module are one or more, in place of or in addition to the functions performed in a particular module. It can be performed in other modules and / or by one or more other devices. Modules may also be implemented across multiple devices and / or other components that are local or remote to each other. Additionally, modules can be moved from one device and added to another, and / or contained within both devices.

As shown in FIG. 1, the contact center 100 may include a central switch 110. The central switch 110 may support an out-of-band connection to a contact via a telecommunications network (not shown) or may receive an incoming contact (eg, caller). The central switch 110 is a contact to help route a contact between one or more contact centers, or to one or more PBX / ACDs or other queues or switching components. It may include routing hardware and software, the contact routing hardware and software includes other internet-based, cloud-based, or other network contact-agent hardware or software-based contact center solutions.

The central switch 110 may not be needed in the contact center system 100, such as when there is only one contact center or when there is only one PBX / ACD routing component. When two or more contact centers are part of the contact center system 100, each contact center may include at least one contact center switch (eg, contact center switches 120A and 120B). The contact center switches 120A and 120B may be communicably coupled to the central switch 110. In embodiments, various topologies of routing and network components can be configured to implement a contact center system.

Each contact center switch for each contact center can be communicatively coupled to multiple (or "group") agents. Each contact center switch may support a number of agents (or "sheets") to be logged in at one time. At any given time, the logged-in agent can be available and can wait to be connected to a contact, or the logged-in agent is connected to another contact or about a call. It may be unavailable for any of many reasons, such as performing some post-call function, such as logging information for, or taking a break.

In the example of FIG. 1, the central switch 110 routes a contact to one of the two contact centers via the contact center switch 120A and the contact center switch 120B, respectively. Each of the contact center switch 120A and the contact center switch 120B is shown with two agents each. Agents 130A and 130B may be logged in to the contact center switch 120A, and agents 130C and 130D may be logged in to the contact center switch 120B.

The contact center system 100 may also be communicably coupled to, for example, integrated services from third party vendors. In the example of FIG. 1, the BP module 140 may be communicably coupled to one or more switches in the switch system of the contact center system 100, such as the central switch 110, the contact center switch 120A, or the contact center switch 120B. In some embodiments, the switch of the contact center system 100 may be communicably coupled to multiple BP modules. In some embodiments, the BP module 140 can be incorporated into a component of a contact center system (eg, incorporated into a switch or "BP switch", or otherwise integrated with a switch or "BP switch". obtain). The BP module 140 refers to a network (eg, the Internet) for agents logged into the switch (eg, agents 130A and 130B), and through another switch (eg, central switch 110), or in some embodiments. Information may be received from a switch (eg, contact center switch 120A) for contacts coming in from (or remote communication networks) (not shown).

The contact center may include multiple pairing modules (eg, BP and FIFO modules) (not shown), and one or more pairing modules may be provided by one or more different vendors. In some embodiments, the pairing module may be a component of the BP module 140, or one or more switches such as the central switch 110 or the contact center switches 120A and 120B. In some embodiments, the BP module may determine which pairing module can handle pairing for a particular contact. For example, the BP module may alternate between enabling pairing via the BP module and enabling pairing with the FIFO module. In other embodiments, one pairing module (eg, a BP module) may be configured to emulate another pairing strategy. For example, a BP module, or a BP component integrated into a BP component in a BP module, may allow the BP module to use BP pairing for a particular contact, or use an emulated FIFO pairing. You can decide if you want to get it. In this case, "BP on" can refer to the time the BP module is applying the BP pairing strategy, and "BP off" means that the BP module is applying a different pairing strategy (eg, FIFO). Can point to the time of.

In some embodiments, single, whether or not the pairing strategies are processed by separate modules, or when several pairing strategies are emulated within a single pairing module. The pairing module of can be configured to monitor and store information about pairings created under any or all pairing strategies. For example, the BP module can observe and record data about the FIFO pairing created by the FIFO module, or the BP module is an emulated FIFO created by the BP module operating in FIFO emulation mode. Data about pairing can be observed and recorded.

FIG. 2 shows an example of a BP payout matrix 200 according to an embodiment of the present disclosure. In this simplified, hypothetical model of a computer-generated contact center system, there are three agents (agents 201, 202 and 203) and three contact types (contact types 211, 212). And 213) exist. Each cell in the matrix represents a "payout" or expected outcome or expected value of a contact-agent interaction between a particular agent and a contact of the indicated contact type. In a real-world contact center system, there can be dozens, hundreds, or more agents, and dozens of contact types, hundreds of contact types, or more. Can exist.

In the BP payout matrix 200, the payout for the dialogue between the agent 201 and the contact of contact type 211 is. 30 or 30%. Other payouts for agent 201 are for contact type 212. 28, for contact type 213. It is 15. The payout for agent 202 is for contact type 211. 30 for contact type 212. 24, for contact type 213. It is 10. The payout for agent 203 is for contact type 211. 25, for contact type 212. 20 for contact type 213. 09.

The payout can represent an expected value for any of a variety of different metrics or optimized variables. Examples of optimized variables include sales conversion rate, customer retention rate, customer satisfaction rate, average processing time measurement, etc., or a combination of two or more metrics. For example, if the BP payout matrix 200 models a retention queue in a contact center system, each payout may represent the possibility that an agent will "save" or retain a particular contact type customer (eg,). The probability that agent 201 will store a contact to be determined for contact type 211 is .30 (or 30%)).

In some embodiments, a model of the contact center system generated by the BP Payout Matrix 200 or other similar computer can be generated by using historical contact-agent dialogue data. For example, the BP payout matrix 200 predicts or otherwise estimates payouts for a given interaction between an agent and a contact type, such as weeks, months, years, etc. of historical data. Rolling windows can be incorporated. If the number of agents changes, the model can be updated to reflect changes in the number of agents, and changes in the number of agents can include hiring new agents, dismissing current agents, etc. Or it involves training current agents in new skills. Contact types can include various types of data, such as information about expected contacts and current customers, such as demographic data and psychographic data, as well as behavioral data such as past purchases or other historical customer information. It can be generated based on relationship management (CRM) data, customer attribute data, third party consumer data, contact center data, etc.). The BP Payout Matrix 200 may be updated in real time to incorporate new contact-agent dialogue data as it becomes available, or it may be updated on a regular basis, such as hourly, nightly, weekly, etc. ..

FIG. 3 represents an example of a simple BP utilization matrix 300 according to an embodiment of the present disclosure. For the BP Payout Matrix 200 (FIG. 2), in this simplified, hypothetical, computer-generated model of the contact center system, there are three agents (agents 201, 202 and 203) and There are three contact types (contact types 211, 212 and 213). In a real-world contact center system, there can be dozens, hundreds, or more agents, and dozens of contact types, hundreds of contact types, or more. Can exist.

Under the BP strategy, the agent is preferably paired with a contact of a particular contact type that follows a computer-generated BP model. In an L1 environment, the contact queue is empty and available to multiple agents, free, or otherwise ready to connect to the contact, to the contact. Waiting for a connection. For example, in a chat context, an agent may have the capacity to chat at the same time as multiple contacts. In these environments, the agent is ready to connect to one or more additional contacts while simultaneously multitasking on one or more other channels such as email and chat. obtain.

In some embodiments, when a contact arrives at a queue or other component of a contact center system, the BP strategy is to determine the type of contact (eg, a contact of contact type 211, 212 or 213). , Analyze information about contacts. The BP strategy determines which agents are available for the connection to the contact, selects the pairing instructions for the most preferred available agents, recommends them, or outputs them otherwise.

In an L2 environment, multiple contacts are waiting in the queue to connect to an agent, one agent is not available, is not available, or is ready to connect to a contact. Not in a state. The BP policy analyzes information about each contact in order to determine the type of each contact (eg, one of contact types 211, 212 or 213 or a contact). In some embodiments, when an agent becomes available, the BP strategy determines which contacts are available for the connection to the agent and selects the pairing instructions for the most preferred available contacts. , Recommended, or output differently.

Each agent has the expected availability or targeted utilization, as shown in the header column of the simple BP utilization matrix 300. In this example, the BP strategy targets balanced 1/3 (“.33”) agent utilization for each of the three agents 201, 202 and 203. Therefore, over time, each agent is expected to be used equally or approximately equally. This configuration of the BP is similar to the FIFO, as long as both the BP and the FIFO target unbiased or balanced agent utilization.

This configuration of the BP is a PBR, as long as performance-based routing (PBR) targets distorted or imbalanced agent usage and intentionally assigns a disproportionate number of contacts to relatively more proven agents. Not similar to. Other configurations of BP can be similar to PBR, as long as other BP configurations can also target distorted agent utilization. Additional information about these features and other features relating to distorting agent or contact utilization (the "copper" or "low" function) is provided, for example, in U.S. Patent Application No. 14 /, incorporated herein by reference. It will be described in Nos. 956,086 and 14/956,074.

As shown in the header column of the simple BP utilization matrix 300, each contact type has expected availability (eg, frequency of incoming calls) or targeted utilization. In this example, a contact of contact type 211 is expected to receive a 50% (“.50”) probability, and a contact of contact type 212 has a 30% (“.30”) probability of receiving. Is expected, and contacts of contact type 212 are expected to receive an incoming call with a remaining 20% (“.20”) probability.

Each cell in the matrix indicates the targeted utilization or expected frequency of contact-agent interactions between a particular agent and the contact of the indicated contact type. In the example of the simple BP utilization matrix 300, agents are expected to be evenly assigned to each contact type according to the frequency of each contact type. Contacts of contact type 211 are expected to arrive in the queue with a 50% probability, and approximately one-third of these contacts are assigned to agents 201, 202, and 203, respectively. Overall, the contact between contact type 211 and agent 201-agent dialogue is expected to occur with a probability of approximately 16% (".16"), and the contact between contact type 211 and agent 202- Agent dialogue is expected to occur with a probability of approximately 16%, and contact-agent dialogue between contact type 211 and agent 203 is expected to occur with a probability of approximately 16%. Similarly, a dialogue between a contact of contact type 212 (30% frequency) and each of the agents 201-203 is expected to occur with a probability of approximately 10% (“.01”), respectively. Dialogue between Type 213 contacts (20% frequency) and each of Agents 201-203 is expected to occur with a probability of approximately 7% (“.07”).

The simple BP utilization matrix 300 also represents approximately the same distribution of contact-agent interactions that would occur under a FIFO pairing strategy, in which each contact-agent dialogue occurs evenly. Will be (normalized for the frequency of each contact type). Under simple BP and FIFO, the targeted (and expected) utilization of each agent is equal (for each of the three agents 201-203, a contact-agent dialogue. One third).

In summary, the BP Payout Matrix 200 (FIG. 2) and the Simple BP Utilization Matrix 300 contact by calculating the average payout weighted according to the frequency distribution of each contact-agent dialogue shown in the Simple BP Utilization Matrix 300. Allows you to determine the expected overall performance of the center system: (.30 + .30 + .25) (.50) (1/3) + (.28 + .24 + .20) (.30) ( 1/3) + (.15 + .10 + .09) (.20) (1/3) ≈ 0.24. Therefore, the expected performance of contact center systems under a simple BP or FIFO is approximately 0.24 or 24%. If the payout represents a retention rate, for example, the expected overall performance would be a retention rate of 24%.

4A-4G show examples of higher performance BP payout matrices and network flows. In this simplified hypothetical contact center, agents or contact types can have different combinations of one or more skills (ie, skill sets), and balanced use across agents and contacts. Linear programming-based network flow optimization techniques can be applied to increase overall contact center performance while maintaining.

FIG. 4A shows an example of a BP skill-based payout matrix 400A according to an embodiment of the present disclosure. The hypothetical contact center system represented in the BP Skill Based Payout Matrix 400A is approximately one-third or. There are three agents (agents 401, 402 and 403), each with 33 expected availability / utilization, and approximately 25% (.15 + .10), 45% (.15 + .30), and. Represented in the BP Payout Matrix 200 (FIG. 2), as long as there are three contact types (contact types 411, 412 and 413) each with an expected frequency / utilization of 30% (.20 + .10) Similar to a contact center system.

However, in this example, each agent is assigned to a particular skill, is trained, or otherwise has a particular skill (or, in other exemplary contact center systems, multiple sets of skills). It has been made available. Examples of skills include a wide range of skills such as technical support, billing support, sales and maintenance, language skills such as English, Spanish and French, "Level 2 Advanced Technical Support", Apple's iPhone®. Includes narrower skills such as technical support for users of Google, technical support for users of Google's Android, and any of a variety of other skills.

Agent 401 is available for contacts that require at least skill 421, agent 402 is available for contacts that require at least skill 422, and agent 403 requires at least skill 423. Available for contacts.

Also in this example, each type of contact can be received in need of one or more of skills 421-423. For example, can a caller to a call center interact with an interactive voice response (IVR) system to determine what skills a particular caller / contact needs for subsequent dialogues. You can interact with the touchtone menu or live with the operator. Another way to consider contact type "skills" is to identify what a contact has, such as buying something from an agent with sales skills or troubleshooting a technical matter with an agent with technical support skills. Needs.

In this example ,. It is expected that 15 or 15% of contacts will be of contact type 411 and will require skill 421 or skill 422. It is expected that 15 or 15% of contacts will be of contact type 412 and will require skills 421 or 422. It is expected that 20 or 20% of contacts will be of contact type 413 and will require skills 421 or 422. It is expected that 10 or 10% of contacts will be of contact type 411 and will require skill 422 or skill 423. It is expected that 30 or 30% of contacts will be of contact type 412 and will require skill 422 or skill 423. It is expected that 10 or 10% of contacts will be of contact type 413 and will require skill 422 or skill 423.

In some embodiments, the agent may have a combination of all skills determined to be required by a particular contact (eg, Spanish skills and iPhone® technical support skills). May be needed. In some embodiments, some skills are preferred but may not be required (ie, no agent with skills for technical support of iPhone® is immediately available or thresholded. If not available for the time, the contact may be paired with an Android technical support agent available instead).

Each cell in the matrix shows the payout of a contact-agent interaction between a particular agent with a particular skill or skill set and a contact with a particular type and need (skill) or set of needs (skill set). .. In this example, the agent 401 with skill 421 can be paired with a contact of any contact type 411, 421 or 413 if the contact of any contact type 411, 421 or 413 requires at least skill 421 ( Have payouts .30, .28 and .15, respectively). Agent 402 with skill 422 can be paired with a contact of any contact type 411, 421 or 413 if a contact of any contact type 411, 421 or 413 requires at least skill 422 (payout, respectively. Has 30, .24, .10, 30, .24 and .10). Agent 403 with skill 423 can be paired with a contact of any contact type 411, 421 or 413 (payout 25, respectively) if the contact of any contact type 411, 421 or 413 requires at least skill 423. , .20 and .09).

Empty cells represent a combination of contacts and agents that would not be paired under this BP pairing strategy. For example, an agent 401 with skill 421 would not be paired with a contact that does not require at least skill 421. In this example, the 18-cell payout matrix contains 6 empty cells and 12 non-empty cells represent 12 pairs of possible pairings.

FIG. 4B shows an example of a BP network flow 400B according to an embodiment of the present disclosure. The BP network flow 400B shows agents 401-403 as "sources" on the left side of the network (or graph) and contact types 411-413 for each skill set as "sinks" on the right side of the network. Each edge in the BP Network Flow 400B represents a possible pairing between a contact and an agent with a particular type and set of needs (skills). For example, edge 401A represents a contact-agent dialogue between agent 401 and a contact of contact type 411 that requires skill 421 or skill 422. Edges 401B, 401C, 402A-F and 403A-C represent other possible contact-agent pairings for each agent and contact type / skill of other possible contact-agent pairings as shown.

FIG. 4C shows an example of a BP network flow 400C according to an embodiment of the present disclosure. The BP network flow 400C is a network / graph display of the BP payout matrix 400A (FIG. 4A). For clarity, the BP Network Flow 400C does not provide an identifier for each edge and instead pays out for each edge for the agent 401 (eg, .30 on edge 401A, .28 on edge 401B, and edge 401C. It is identical to BP Network Flow 400B (FIG. 4B), except that it shows the corresponding payout for each edge for agents 402 and 403, as well as .15) above.

FIG. 4D shows an example of a BP network flow 400D according to an embodiment of the present disclosure. For clarity, the BP Network Flow 400D does not show skills for each agent 401-403 and is instead required by the relative "supply" provided by each agent and the combination of each contact type / skill. It is the same as the BP network flow 400C (FIG. 4C) except that it indicates the "request" to be made. Each agent provides a "supply" equal to each agent's expected availability or target utilization (one-third each for a total supply of 1 or 100%). Each contact type / skill has an amount of agent supply equal to the expected frequency or target utilization of each contact type / skill (for a total request of 1 or 100%, .15, .15, .20, .10, respectively. , .30, and .10) are required. In this example, the total supply and demand are normalized or otherwise configured to be equal to each other, and the capacity or bandwidth along each edge is considered to be infinite or otherwise unlimited. (That is, the edge can explain "who can be paired with whom" rather than "how much" or "how many times"). In other embodiments, there may be supply / demand imbalances, or there may be quotas set for some or all edges or otherwise limited capacity.

FIG. 4E shows an example of a BP network flow 400E according to an embodiment of the present disclosure. The BP Network Flow 400E is identical to the BP Network Flow 400D (FIG. 4D), except that supplies and requirements are scaled at a magnification of 3000 for ease of expression. When scaling, the supply for each agent is shown as 1000 instead of one-third, and the total supply is shown as 3000. Similarly, the relative requirements for each contact type / skill set are scaled, as well as a total of 3000. In some embodiments, scaling does not occur. In other embodiments, the amount of scaling can vary and can be greater than or less than 3000.

FIG. 4F shows an example of a BP network flow 400F according to an embodiment of the present disclosure. BP Network Flow 400F with BP Network Flow 400E (Fig. 4E), except that for clarity, payouts along each edge are not shown and instead show one solution for BP Network Flow 400F. It is the same. In some embodiments, a "maximum flow" or "max flow" algorithm, or other linear programming algorithm, is a supply (source) to meet a requirement (sink) that can balance the use of agents and contacts. Can be applied to the BP Network Flow 400F to determine one or more solutions for optimizing the "flow" or "distribution" of.

In some embodiments, the goal may also be to maximize the overall expectation of the metric or the metric being optimized. For example, in the sales queue, the metric for optimization can be the conversion rate, and the max flow goal is to maximize the overall expected conversion rate. In an environment where multiple Maxflow solutions are available, one technology to choose a solution is a "Maximum Cost" or "MaxCost" solution (ie, a solution that offers the highest overall payoff under Maxflow. ) Can be selected.

In this example, agents 401-403 represent sources and contact types 411-413 with different skill set combinations represent sinks. In some contact center environments, such as L2 (over-contact) environments, network flow is the source from which the contacts waiting in the queue provide the supply, and the sink where the possible agents that may be available provide the request. Can be reversed, as is.

BP Network Flow 400F presents the optimal flow solution determined by the BP module or similar components. According to this solution (there may be some solutions to choose from, or some randomly selected solutions), the edge 401A (from agent 401 to contact type 411 with skills 421 and 422) Edge 401B (from agent 401 to contact type 412 with skills 421 and 422) having an optimal flow of 0 has an optimal flow of 400 and (from agent 401 to skills 421 and 422). Edge 401C (to contact type 413 with) has an optimal flow of 600. Similarly, the optimal flows for edges 402A-F for agent 402 are 450, 50, 0, 300, 200, and 0, respectively, and the optimal flows for edges 403A-C for agent 403 are 0, respectively. 700 and 300. As described in detail below, this optimal flow solution also maximizes the expected overall performance of the contact center system according to payouts for each pair of agent and contact type / skill set, while also maximizing the agent and Describe the relative percentage of contact-agent interactions (or the relative likelihood of choosing a particular contact-agent dialogue) to achieve targeted utilization of contacts.

FIG. 4G shows an example of a BP network flow 400G according to an embodiment of the present disclosure. The BP Network Flow 400G is identical to the BP Network Flow 400F (FIG. 4F), except that for clarity, the edge for which the optimal flow solution was determined to be 0 has been removed. Under BP strategy, agents will not preferably be paired with a contact type whose optimal flow solution has been determined to be 0, even with supplementary skills and non-zero payouts.

In this example, edges 401A, 402C, 402F and 403A have been deleted. The remaining edges represent the preferred pairing. Thus, in an L1 (agent excess) environment, when a contact arrives, the contact can preferably be paired with one of the agents for which preferred pairing is available. For example, a contact of contact type 411 with skills 421 and 422 requires 450 units of total supply (availability) of agent 402 and can always be preferably paired with agent 402. For another example, contacts of contact type 412 with skills 421 and 422 can preferably be paired with agent 401 from time to time and sometimes with agent 402. This contact has a total request of 450 (based on the expected frequency of incoming contacts / skills of this type), requesting a supply of 400 units from Agent 401 and a supply of the remaining 50 units from Agent 402. ..

In some embodiments, when a contact of this type / skill arrives, the BP module or similar component is either Agent 401 or 402 according to the relative requirements (400 and 50) made by each agent. Can be selected. For example, a pseudo-random number generator can be used to randomly select agents 401 or 402 by random selection weighted according to relative requirements. Therefore, for each contact of this type / skill, the probability of selecting agent 401 as the preferred pairing is 400/450 (≈89%), and the probability of selecting agent 402 as the preferred pairing is 50/450 (≈89%). ≈ 11%). Over time, if many contacts of this type / skill are paired using BP strategies, approximately 89% of those contacts can preferably be paired with Agent 401 and of those contacts. The remaining 11% of which can be preferably paired with the agent 402. In some embodiments, the agent's overall target utilization, or target utilization by contact type / skill, can be the agent's "bandwidth" for receiving a proportional percentage of contacts.

For this BP Network Flow 400G with this solution, the total supply from all of the agents is expected to meet the total requirements from all of the contacts. Therefore, targeted utilization of all agents (here balanced utilization) is expected to be higher in the contact center system according to the payouts and the relative distribution of agents to contacts along the edges with those payouts. Can be achieved while also achieving the overall performance achieved.

5A-I show examples of different BP payout matrices and network flows. For some configurations of contact types and agents with various combinations of skills, the optimal flow or max flow for a given BP network flow may not be perfectly balanced between supply and demand. This example is similar to the example in FIGS. 4A-4G, except that this configuration of agents and contact types initially leads to disproportionate supply and demand. In this simplified, hypothetical contact center, quadratic programming-based techniques for coordinating target utilization traverse agents and contacts to address an imbalanced configuration of agents and contact types. It can be applied in conjunction with linear programming-based network flow optimization techniques to increase overall contact center performance while maintaining optimally distorted utilization.

FIG. 5A represents an example of a BP skill-based payout matrix 500A according to an embodiment of the present disclosure. The hypothetical contact center systems represented in the BP Skill Based Payout Matrix 500A are approximately one-third or more, respectively. The contact center represented in the BP Skill Based Payout Matrix 400A (FIG. 4), as long as there are 33 agents (agents 501, 502 and 503) with initially expected availability / utilization Similar to the system. There are two contact types (contact types 511 and 512) with expected frequencies / utilization of approximately 40% (.30 + .10) and 60% (.30 + .30), respectively.

In this example, agents 501 are assigned, trained, or otherwise made available for skills 521 and 522, and agents 502 and 503 are assigned or trained only for skills 522. It has been or has been made available separately. For example, if skill 521 represents a French language skill and skill 522 represents a German language skill, agent 501 may be assigned to a contact that requires either French or German, but agent 502 and 503 can only be assigned to contacts that require German, not contacts that require French.

In this example ,. It is expected that 30 or 30% of contacts will be of contact type 511 and will require skill 521. It is expected that 30 or 30% of contacts are of contact type 512 and require skill 521. It is expected that 10 or 10% of contacts are of contact type 511 and require skill 522. It is expected that 30 or 30% of contacts are of contact type 512 and require skill 522.

In this example, Agent 501 can be paired with any contact (having payouts .30, .28, .30 and .28, as shown in the BP Skill Based Payout Matrix 500A). Agents 502 and 503 with only skill 522 require contact types 511 and 512 to have at least skill 522 (having payouts .30, .24, .25 and .20 as shown in the BP skill-based payout matrix 500A). If so, it can be paired with a contact of either contact type 511 or 512. Agents 502 and 503 will not be paired with contacts that require only skill 521, as indicated by the empty cells in the BP skill-based payout matrix 500A. The 12-cell payout matrix contains 4 empty cells, with 8 non-empty cells representing 8 pairs of possible pairings.

FIG. 5B shows an example of a BP network flow 500B according to an embodiment of the present disclosure. Similar to BP Network Flow 400B (FIG. 4B), BP Network Flow 500B shows agents 501-503 as sources on the left side of the network and contact types 512 and as sinks for each skill set on the right side of the network. 5123 is shown. Each edge in the BP Network Flow 500B represents a possible pairing between a contact and an agent with a particular type and set of needs (skills). Edges 501A-D, 502A-B and 503A-B represent possible contact-agent pairings for each agent and contact type / skill of possible contact-agent pairings as shown.

FIG. 5C shows an example of a BP network flow 500C according to an embodiment of the present disclosure. The BP network flow 500C is a network display of the BP payout matrix 500A (FIG. 5A). For clarity, no identifier is given for each edge, instead payouts for each edge (eg, .30 on edges 501A and 501D, and .28 on edges 501B and 501C), and agents 502 and 503. Shows the corresponding payout for each edge.

FIG. 5D shows an example of a BP network flow 500D according to an embodiment of the present disclosure. The BP Network Flow 500D shows the relative initial supply provided by each agent and the requirements required by each contact type / skill combination. The total supply 1 is equal to the total request 1.

FIG. 5E shows an example of a BP network flow 500E according to an embodiment of the present disclosure. For ease of expression, supplies and requirements are scaled at a magnification of 3000 and a max flow solution for the initial supply is shown for each edge. According to this solution, edge 501A (from agent 501 to contact type 511 with skill 521) has an optimal flow of 900 and edge 501B (from agent 501 to contact type 512 with skill 521). Has an optimal flow of 100, and edges 501C and 501D have an optimal flow of 0. Similarly, the optimal flows for agent 502 are 300 and 700, respectively, and the optimal flows for agent 503 are 0 and 200, respectively.

According to this solution, Agent 503 is substantially underutilized as compared to Agents 501 and 502. Agents 501 and 502 are each optimized for a complete supply of their 1000 units, but Agent 503 is expected to use only 200 units or only one-fifth of the supply of Agent 503. Be done. In a contact center environment, the agent 503 can be assigned to fewer contacts and spend more free time compared to agents 501 and 502, or the agent can be assigned to unfavorable contacts. , Brings lower contact center performance than expected by Maxflow Solutions.

Similarly, according to this solution, contacts of contact type 512 that require skill 521 are substantially underutilized (or "insufficient service" compared to other contact type / skill combinations. "). Other contact type / skill combinations are optimized for the full requirements of their 900, 300, and 900 units, respectively, but the contact type 512 that requires skill 521 is only 100 units or Or it is expected to receive only one-ninth of the requirements for this contact type / skill. In a contact center environment, this underutilized contact type / skill combination may experience longer latency compared to other contact type / skill combinations, or the contact may be assigned to an unfavorable agent. It can result in lower contact center performance than expected by Maxflow Solutions.

The solution presented in BP Network Flow 500E still balances total supply and demand, but agent 503 can be selected much less frequently than other agents and / or some contacts are more. You may have to wait for a preferred agent for a long time and / or the overall contact center performance may not achieve the overall payout expected by the Max Flow solution.

5F and 5G show relative agents to improve the balance of agent and contact utilization in the contact center system when the Maxflow solution is imbalanced, as in the BP Network Flow 500E (Figure 5E). The techniques of some embodiments for adjusting the supply are shown.

FIG. 5F shows an example of a BP network flow 500F according to an embodiment of the present disclosure. In the BP Network Flow 500F, agents sharing the same skill set are "grouped" into a single network node. In this example, agents 502 and 503 are combined into a single node for skill 522 with 2000 combined total supplies.

Similarly, contact types that share the same skill set are grouped together on a single network node. In this example, contact types 511 and 512 requiring skill 521 are combined into a single node for skill 521 with 1800 combined aggregate requirements, and contact types 511 and 512 requiring skill 522. Is combined into a single node for skill 522 with 1200 combined total requirements.

In addition, the edges are grouped together. For example, the four edges emanating from agents 502 and 503 (edges 502A, 502B, 503A and 503B, as labeled in FIG. 5B) require skill 522 from a "supernode" for agents with skill 522. It is grouped into a single edge that emits to the "supernode" for the contact type.

At this point, in some embodiments, a quadratic algorithm or similar technique may be applied to the combined network to adjust the relative supply of agents.

FIG. 5G shows an example of a BP network flow 500G according to an embodiment of the present disclosure. The BP Network Flow 500G exhibits an agent supply tuned according to a solution for a quadratic algorithm or similar technology. In this example, the supply for agent supernodes for skills 521 and 522 is adjusted from 1000 to 1800 in the BP network flow 500F (FIG. 5F), and the supply for agent supernodes for skills 522 is. It is adjusted from 2000 to 1200 in the BP network flow 500F.

The total supply can remain the same (eg 3000 in this example), but the relative supply for agents of various skill sets has been adjusted. In some embodiments, the total supply for a single supernode can be evenly distributed among the agents within the supernode. In this example, the supply of 1200 units for the agent supernode to skill 522 is evenly divided among the agents, distributing 600 units to each of agents 502 and 503.

FIG. 5H shows an example of a BP network flow 500H according to an embodiment of the present disclosure. BP Network Flow 500H represents a Max Flow solution calculated by using the tuned supply shown in BP Network Flow 500G (FIG. 5G). Agent 501 has a 1800 tuned supply, agent 502 has 600 tuned supplies, and agent 503 has 600 tuned supplies. According to this solution, the edge 501A (from agent 501 to contact type 511 with skill 521) still has an optimal flow of 900 and the edge (from agent 501 to contact type 512 with skill 521). 501B currently has an optimal flow of 900, and edges 501C and 501D still have an optimal flow of 0. Similarly, the optimal flows for agent 502 are currently 300 and 300, respectively, and the optimal flows for agent 503 are currently 0 and 600, respectively.

FIG. 5I shows an example of a BP network flow 500I according to an embodiment of the present disclosure. The BP Network Flow 500I is identical to the BP Network Flow 500H, except that for clarity, the edge where the optimal flow solution was determined to be 0 has been removed. In this example, the edges 501C, 501D and 503A have been removed.

Using the solutions shown in BP Network Flows 500H and 500I, all contact type / skill combinations can be fully utilized (“service is complete”).

In addition, overall agent utilization can be more balanced. Under BP Network Flow 500E (FIG. 5E), Agent 503 can only utilize one-fifth of Agents 501 and 502. Thus, agents 501 and 502 will each be assigned approximately 45% of contacts, while agent 503 will be assigned only the remaining approximately 10% of contacts. Under BP Network Flow 500H, Agent 501 may be assigned approximately 60% of contacts, and Agents 502 and 503 may each be assigned the remaining approximately 20% of contacts. In this example, the busiest agent (Agent 501) will receive only three times as many contacts as the least busy agent, instead of receiving five times as many contacts as the least busy agents (Agents 502 and 503). ..

FIG. 6 represents a flow diagram of the BP skill-based payout matrix method 600 according to an embodiment of the present disclosure. The BP skill-based payout matrix method 600 may start at block 610.

At block 610, historical contact-agent result data can be analyzed. In some embodiments, a rolling window of historical contact-agent result data, such as a weekly window, a month window, a ninety day window, or a year window, can be analyzed. Historical Contact-Agent result data is a separate interaction between a contact and an agent, including which agent communicated with which contact, when the communication occurred, the duration of the communication, and an identifier for the result of the communication. May contain information about. For example, in a telemarketing call center, the result may indicate whether or not the sale has occurred, or the amount of the sale if the sale has occurred. In the customer retention queue, the result may indicate whether the customer was retained (“saved”) or the amount of any incentive offered to retain the customer. In the customer service queue, the results are whether the customer's needs are met or the problem is resolved, or a score (eg, Net Promoter Score or NPS) or other assessment that represents customer satisfaction with the contact-agent dialogue. Can be shown. After or in parallel with the historical contact-agent result data analysis, the BP skill-based payout matrix method 600 may proceed to block 620.

At block 620, contact attribute data can be analyzed. Contact attribute data may include data stored in one or more customer relationship management (CRM) databases. For example, the CRM database of a wireless communication provider shows the type of mobile phone used by the customer, the type of contract the customer has applied for, the duration of the customer's contract, the monthly fee of the customer's contract, and the relationship between the customer and the company. May include information about retention period. For another example, a bank's CRM database may contain information about the type and number of accounts held by the customer, the average monthly balance of the customer's account, and the duration of the customer's relationship with the company. In some embodiments, the contact attribute data may also include third party data stored in one or more databases obtained from a third party. After or in parallel with the analysis of contact attribute data, the BP skill-based payout matrix method 600 can proceed to block 630.

At block 630, skill groups may be determined for each agent and each contact type. Examples of skills include a wide range of skills such as technical support, billing support, sales and maintenance, language skills such as English, Spanish and French, "Level 2 Advanced Technical Support", Apple's iPhone®. Includes narrower skills such as technical support for users of Google, technical support for users of Google's Android, and any of a variety of other skills. In some embodiments, any particular skill may not exist, or only one skill may be identified across all of the agents or all of the contact types. In these embodiments, there can only be a single "skill group".

In some embodiments, a given contact type may require different skill sets at different times. For example, during the first call to the call center, one type of contact may have a technical question and may require an agent with technical support skills, but during the second call, the same type of contact. The same contact may have a billing question and may require an agent with customer support skills. In these embodiments, the same contact type may be included more than once, depending on each contact type / skill combination. After the skill group is determined, the BP skill-based payout matrix method 600 may proceed to block 640.

At block 640, target utilization can be determined for each agent, and anticipation rates can be determined for each contact type (or contact type / skill combination). In some L1 environments, balanced agent utilization can be targeted so that each agent is expected to be assigned an approximately equal number of contacts over time. For example, if the contact center environment has four agents, each agent may have 1/4 (or 25%) targeted utilization. For another example, if the contact center environment has n agents, each agent may have 1 / n (or equal percentage of contacts) targeted utilization.

Similarly, the expected rate for each contact type / skill can be determined, for example, based on the actual rate observed in the historical contact-agent result data analyzed in block 610. After the target utilization and anticipation rate have been determined, the BP skill-based payout matrix method 600 can proceed to block 650.

At block 650, a payout matrix with the expected payout for each viable contact-agent pairing can be determined. In some embodiments, contact-agent pairing may be feasible if the agent and contact type have at least one common skill. In other embodiments, contact-agent pairing may be feasible if the agent has at least all of the skills required by the contact type. In yet other embodiments, other heuristics for feasibility may be used.

An example of a payout matrix, which is a BP skill-based payout matrix 400A, is described in detail above with reference to FIG. 4A. The BP Skill Based Payout Matrix 400A has an expected frequency determined based on a set of agents with associated skills and targeted utilization, historical contact-agent result data and / or contact attribute data (various skills). Includes a set of contact types (combined with the set) and a set of non-zero expected payouts for each viable contact-agent pairing. After the payout matrix is determined, the BP skill-based payout matrix method 600 can proceed to block 660.

At block 660, the model generated by the computer processor can be output according to the payout matrix. For example, a component such as a computer processor embedded within a contact center system or a computer processor communicatively coupled to a contact center system, or a BP module in a contact center system is another component of the contact center system or computer processor. Can output the payout matrix model received by. In some embodiments, the payout matrix model can be logged, printed, displayed, transmitted, or otherwise to other components of the contact center system or to a human administrator. Can be remembered like. After the payout matrix model is output, the BP skill-based payout matrix method 600 may be terminated.

FIG. 7A shows a flow chart of the BP network flow method 700A according to an embodiment of the present disclosure. The BP network flow method 700A may start at block 710.

At block 710, the BP payout matrix can be determined. In some embodiments, the BP payout matrix can be determined using the BP payout matrix method 600 or a similar method. In other embodiments, the BP payout matrix may be received from another component or module. After the BP payout matrix is determined, the BP network flow method 700A can proceed to block 720.

At block 720, target utilization can be determined for each agent and anticipation rate can be determined for each contact type. In another embodiment, the payout matrix determined in block 710 has a targeted utilization and / or expected rate, such as the payout matrix output by the BP payout matrix method 600, or the BP skill-based payout matrix 400A (FIG. 4A). It can be incorporated or otherwise included. After the target utilization and anticipation rate have been determined, the BP network flow method 700A may proceed to block 730 if necessary.

At block 730, agent supply and contact type requirements can be determined. In the above, as described in detail, for example with reference to FIGS. 4D and 4E, each agent has an expected availability or targeted utilization of each agent (eg, in an environment with three agents, 1). Or, for every 100% total supply, each can provide a "supply" equal to one-third). In addition, each contact type / skill may request an amount of agent supply equal to the expected frequency or target utilization of each contact type / skill for a 1 or 100% total request. Total supplies and requirements can be normalized or configured differently to be equal to each other, and the capacity or bandwidth along each edge can be considered infinite or otherwise unlimited. In other embodiments, there may be supply / demand imbalances, or there may be quotas set for some or all edges or otherwise limited capacity.

In some embodiments, supplies and requirements can be scaled by several magnifications, such as 1000, 3000. When scaling, the supply for each of the three agents can be shown as 1000 instead of one-third, and the total supply can be shown as 3000. Similarly, the relative requirements for each contact type / skill set can be scaled. In some embodiments, scaling does not occur. After the agent supply and contact type requirements have been determined, the BP network flow method 700A may proceed to block 740.

At block 740, the preferred contact-agent pairing can be determined. In the above, one or more solutions for BP network flow can be determined, for example, as described in detail with reference to FIGS. 4F and 4G. In some embodiments, the "maximum flow" or "max flow" algorithm, or other linear programming algorithm, optimizes the "flow" or "distribution" of the supply (source) to meet the demand (sink). It can be applied to the BP network flow to determine one or more solutions to do so. In some embodiments, the "max cost" algorithm can be applied to select the optimal max flow solution.

In some contact center environments, such as L2 (over-contact) environments, network flow is the source from which the contacts waiting in the queue provide the supply, and the sink where the possible agents that may be available provide the request. Can be reversed, as is.

The BP network flow may include an optimal flow solution determined by the BP module or similar components. According to this solution (there may be some solutions to choose from, or some randomly selected solutions), some (feasible) edges can have an optimal flow of 0, It indicates that such feasible pairing is not the preferred pairing. In some embodiments, the BP network flow may remove an edge that represents a feasible pairing if it is determined that the pairing is not the preferred pairing.

Other edges may have a non-zero optimal flow, which indicates that such feasible pairing is at least sometimes preferable. As explained in detail above, this optimal flow solution also maximizes the expected overall performance of the contact center system according to the payout for each pair of agent and contact type / skill set, while also maximizing the agent and Describe the relative percentage of contact-agent interactions (or the relative likelihood of choosing a particular contact-agent dialogue) to achieve targeted utilization of contacts.

For some solutions of some BP network flows, a single contact type / skill can have multiple edges entering a single contact type / skill from multiple agents. In these environments, the contact type / skill can have multiple preferred pairings. Given the selection from multiple agents, the BP network flow provides a relative percentage or weight that one of several agents can be selected each time a contact of that contact type / skill arrives at the contact center. Shown. After determining the preferred contact-agent pairing, the BP network flow method 700A can proceed to block 750.

At block 750, the model generated by the computer processor may be output according to the preferred contact-agent pairing. For example, a component such as a computer processor embedded within a contact center system or a computer processor communicatively coupled to a contact center system, or a BP module in a contact center system is another component of the contact center system or computer processor. The preferred pairing model received by can be output. In some embodiments, the preferred pairing model can be logged, printed, displayed, transmitted, or sent to other components of the contact center system or to a human administrator. Can be remembered differently. After the preferred pairing model is output, the BP network flow method 700A can be terminated.

FIG. 7B shows a flow chart of the BP network flow method 700B according to the embodiment of the present disclosure. The BP network flow 700B is similar to the BP network flow 700A described above with reference to FIG. 7A. The BP network flow method 700B may start at block 710. At block 710, the BP payout matrix can be determined. After the BP payout matrix is determined, the BP payout matrix method 700B can proceed to block 720. At block 720, target utilization can be determined for each agent and anticipation rate can be determined for each contact type. After the target utilization and anticipation rate have been determined, the BP network flow method 700B may proceed to block 730 if necessary. At block 730, agent supply and contact type requirements can be determined. After the agent supply and contact type requirements have been determined, the BP network flow method 700B may proceed to block 735.

At block 735, the agent supply and / or contact request may be adjusted to balance the agent utilization or improve the agent utilization equilibrium. In the above, agents sharing the same skill set can be "grouped" into a single network node (or "supernode"), for example, as described in detail above with reference to FIGS. 5F and 5G. .. Similarly, contact types that share the same skill set can be grouped into a single network node. In addition, the edges can be grouped according to the corresponding supernodes of the edges. At this point, in some embodiments, a quadratic algorithm or similar technique has been applied to the combined network to coordinate the relative supply of agents and / or the relative requirements of contacts. obtain. After adjusting the agent supply and / or contact request to balance agent utilization, BP network flow method 700B may proceed to block 740.

At block 740, the preferred contact-agent pairing can be determined. After determining the preferred contact-agent pairing, the BP network flow method 700A can proceed to block 750. At block 750, the model generated by the computer processor may be output according to the preferred contact-agent pairing. After the preferred pairing model is output, the BP network flow method 700B can be terminated.

FIG. 8 shows a flow chart of the BP network flow method 800 according to an embodiment of the present disclosure. The BP network flow method 800 may start at block 810.

At block 810, available agents may be determined. Dozens of agents, hundreds of agents, or more can be employed in the real-world queue of contact center systems. At any given time, a small number of these hired agents may be logged into the system or otherwise active in shifts. In addition, at any given time, a small number of logged-in agents may engage in contact interactions (eg, in a call to a call center) or log the results of recent contact interactions. It may be available to get, take a break, or otherwise be assigned to an incoming contact. The rest of the logged-in agent can be free or available to be assigned differently. After determining the set of agents available, BP network flow method 800 may proceed to block 820.

At block 820, a preferred contact-agent pairing BP model can be determined. In some embodiments, the preferred pairing model can be determined by using the BP network flow method 700A (FIG. 7A) or 700B (FIG. 7B), or a similar method. In other embodiments, the preferred pairing model may be received from another component or module.

In some embodiments, the preferred pairing model may include all agents employed for a contact center queue or multiple queues. In other embodiments, the preferred pairing model may include only agents logged in to the queue at a given time. In yet other embodiments, the preferred pairing model may include only agents determined to be available in block 810. For example, referring to FIG. 4B, if agent 403 is unavailable, some embodiments differ, omitting nodes for agent 403 and including nodes for agents 401 and only agents 402 that are available. A preferred pairing model can be used. In other embodiments, the preferred pairing model may include a node for the agent 403, but may be adapted to avoid generating a non-zero probability of assigning contacts to the agent 403. For example, the flow capacity from agent 403 to each compatible contact type can be set to zero.

Preferred pairing models can be pre-computed (eg, retrieved from cache or other storage), or when agents become available or unavailable, and / or with various skill needs. When a type of contact arrives at the contact center, it can be calculated in real time or near real time. After the preferred pairing model has been determined, the BP network flow method 800 can proceed to block 830.

At block 830, the available contacts can be determined. For example, in an L1 environment, multiple agents are available, waiting for contact assignments, and the contact queue is empty. When a contact arrives at the contact center, the contact can be assigned to one of the available agents without having to wait in hold. In some embodiments, the preferred pairing model determined in block 820 can be determined first or updated after the determination of available contacts in block 830. For example, referring to FIGS. 4A-4G, agents 401-403 can be three agents out of dozens or more that happen to be available at a given moment. The BP skill-based payout matrix 400A (FIG. 4A) can then be determined for the three agents available at that time, and the BP network flow 400G (FIG. 4G) is based on the BP skill-based payout matrix 400A. It can be determined about the three agents available at that time. Therefore, a preferred pairing model can then be determined for the three available agents.

In some embodiments, the particular contact to be paired is the preferred pairing model, even though it is already known in BP Network Flow Method 800, since that contact has already been determined in block 830. , For example, may occupy some or all of the expected contact type / skill combinations in the BP Network Flow 400G. After the available contacts have been determined in block 830 (and in some embodiments a preferred pairing model has been generated or updated), the BP network flow method 800 can proceed to block 840.

At block 840, at least one preferred contact-agent pairing of available agents and available contacts can be determined. For example, as shown in BP Network Flow 400G (FIG. 4G), if the available contacts are of contact type 411 and require skill 421 or 422, a preferred pairing is pairing with agent 402. It is a ring. Similarly, if the available contacts are of contact type 412 and require skill 421 or 422, the preferred pairing is Agent 401 (optimal flow of 400) or Agent 402 (optimal flow of 50). ) Is paired with any of them. After at least one preferred contact-agent pairing has been determined, the BP network flow method 800 can proceed to block 850.

At block 850, one of at least one preferred contact-agent pairing may be selected. In some embodiments, the choice can be random, such as by using a pseudo-random number generator. The likelihood (or probability) of choosing a given one of at least one preferred contact-agent pairing can be based on the statistical possibilities described by the BP model. For example, as shown in BP Network Flow 400G (FIG. 4G), if the available contacts are of contact type 412 and require skill 421 or 422, the probability of choosing agent 401 is 400 /. The probability of selecting the agent 402 is 450 ≈ 89%, and the probability of selecting the agent 402 is 50/450 ≈ 11%.

In some embodiments, the need for random selection cannot exist, as the selection can be self-explanatory if there is only one preferred contact-agent pairing. For example, as shown in BP Network Flow 400G (FIG. 4G), if the available contacts are of contact type 411 and require skill 421 or 422, the preferred pairing is always with agent 402. It is pairing, and the probability of selecting the agent 402 is 450/450 ≈ 100%. After selecting one of at least one preferred contact-agent pairing, the BP network flow method 800 may proceed to block 860.

At block 860, the selected pairing may be output for connection in the contact center system. For example, a component such as a computer processor embedded within a contact center system or a computer processor communicatively coupled to a contact center system, or a BP module in a contact center system is another component of the contact center system or computer processor. It may output the preferred pairing selection (or recommended pairing, or pairing instruction) received by. In some embodiments, the preferred pairing selection can be logged, printed, displayed, transmitted, or sent to other components of the contact center system or to a human administrator. Can be remembered differently. The receiving component may use the preferred pairing selection to connect the selected agent to the contact for which pairing was requested or otherwise determined. After the preferred pairing instruction is output, the BP network flow method 800 can be terminated.

FIG. 9 shows a flow chart of the BP network flow method 900 according to an embodiment of the present disclosure. In some embodiments, the BP network flow method 900 is similar to the BP network flow method 800. The BP network flow method 800 exemplifies the L1 environment (excessive agents), while the BP network flow method 900 exemplifies the L2 environment (contacts are in the queue). The BP network flow method 900 may start at block 910.

At block 910, available contacts can be determined. Dozens of agents, hundreds of agents, etc. can be employed in the real-world queue of contact center systems. In the L2 environment, all logged-in agents are either engaged in contact dialogue or otherwise unavailable. When a contact arrives at the contact center, the contact may be asked to wait in the hold queue. At a given time, dozens or more contacts can wait in hold. In some embodiments, the queue can be ordered sequentially by incoming time, with the longest waiting contact being ordered at the top of the queue. In other embodiments, the queue may be ranked at least partially based on the priority rating or the state of the individual contacts. For example, a contact designated as "higher priority" may be positioned at or near the beginning of the queue ahead of other "normal priority" contacts that have been waiting longer. After determining the set of available contacts waiting in the queue, BP network flow method 900 may proceed to block 920.

At block 920, a preferred contact-agent pairing BP model can be determined. In some embodiments, the preferred pairing model is the BP network flow method 700A (FIG. 7A), as long as the waiting contact provides the source of supply and the potentially available agent provides the sink of the request. ) Or by using a method similar to 700B (FIG. 7B). In other embodiments, the preferred pairing model may be received from another component or module.

In some embodiments, the preferred pairing model may include all contact types that are expected to arrive in the contact center queue or multiple queues. In other embodiments, the preferred pairing model may include only contact type / skill combinations that are in the queue and are waiting in the queue when the model is requested. For example, consider a contact center system that expects to receive three types X, Y and Z contacts, but only type X and Y contacts are currently waiting in the queue. Some embodiments may use different preferred pairing models that include only nodes for waiting contacts of types X and Y and omit nodes for contact type Z. In other embodiments, a preferred pairing model may include a node for contact type Z, but this model may be adapted to avoid generating a non-zero probability of assigning an agent to a contact of contact type Z. .. For example, the flow capacity from contact type Z to each adaptable agent can be set to zero.

Preferred pairing models can be pre-computed (eg, retrieved from cache or other storage), or when agents become available or unavailable, and / or with various skill needs. When a type of contact arrives at the contact center, it can be calculated in real time or near real time. After the preferred pairing model has been determined, the BP network flow method 900 can proceed to block 930.

At block 930, available agents may be determined. For example, in an L2 environment, multiple contacts are waiting, available for assignment to agents, and all agents can be occupied. When an agent becomes available, the agent can be assigned to one of the waiting contacts without remaining free. In some embodiments, the preferred pairing model determined in block 920 can be determined first or updated after the determination of available contacts in block 830. For example, there can be three contacts waiting in a queue, each with a different skill and type of contact. The BP skill-based payout matrix can be determined for the three contacts waiting at that time, and the BP network flow is determined for the three contacts waiting at that time based on the BP skill-based payout matrix. Can be done. Therefore, the preferred pairing model can then be determined for the three waiting contacts.

In some embodiments, the preferred pairing model is latent, even though the particular agent to be paired is already known to the BP network flow method 900, since the agent has already been determined in block 930. Can occupy some or all of the available agents. The BP network flow method 900 may proceed to block 940 after the available agents have been determined in block 930 (and in some embodiments a preferred pairing model has been generated or updated).

At block 940, at least one preferred contact-agent pairing of available agents and available contacts can be determined. After at least one preferred contact-agent pairing has been determined, the BP network flow method 900 can proceed to block 950.

At block 950, one of at least one preferred contact-agent pairing may be selected. In some embodiments, the choice can be random, such as by using a pseudo-random number generator. The likelihood (or probability) of choosing a given one of at least one preferred contact-agent pairing can be based on the statistical possibilities described by the BP model. In some embodiments, the need for random selection cannot exist, as the choice can be self-explanatory if there is only one preferred contact-agent pairing. After selecting one of at least one preferred contact-agent pairing, the BP network flow method 900 may proceed to block 960.

At block 960, the selected pairing may be output for connection in the contact center system. For example, a component such as a computer processor embedded within a contact center system or a computer processor communicatively coupled to a contact center system, or a BP module in a contact center system is another component of the contact center system or computer processor. It may output the preferred pairing selection (or recommended pairing, or pairing instruction) received by. In some embodiments, the preferred pairing selection can be logged, printed, displayed, transmitted, or sent to other components of the contact center system or to a human administrator. Can be remembered differently. The receiving component may use the preferred pairing selection to connect the selected agent to the contact for which pairing was requested or otherwise determined. After the preferred pairing instruction is output, the BP network flow method 900 can be terminated.

In some embodiments, the BP payout matrix and network flow model can be used in an L3 environment (ie, multiple available agents and multiple contacts waiting in a queue). In some embodiments, the network flow model can be used to batch pair multiple contact-agent pairings at the same time. BP pairing in an L3 environment is described in detail, for example, in US Patent Application No. 15 / 395,469, incorporated herein by reference. In other embodiments, if the contact center system is operating in an L1 and / or L2 environment, a BP network flow model can be used, and if the contact center system is operating in an L3 (or L0) environment, an alternative. BP pairing strategy can be used.

In the example described above, the BP network flow model targets agent utilization that is balanced (or as close as possible to a viable state for a particular contact center environment). .. In other embodiments, distorted agent utilization or otherwise imbalanced agent utilization can be targeted (eg, "copper" technology) and / or distorted contact utilization or otherwise imbalanced. Contact utilization can be targeted (eg, "low" technology). Examples of these techniques, including copper and row techniques, are described in detail in, for example, U.S. Patent Applications 14/956,086 and 14/956,074 above, which are described herein. Incorporated by reference.

How many embodiments, such as an embodiment in which a BP module (eg, BP module 140) is fully integrated within a contact center switch (eg, central switch 110, contact center switch 120A, etc.) or otherwise integrated within a contact center switch. In such an embodiment, the switch may implement the BP technique without separating the pairing request and response between the switch and the BP module. For example, the switch may determine its own cost function or multiple functions to apply to each possible pairing when the need arises, and the switch will automatically minimize (or or) cost features as appropriate. In some embodiments, it can be maximized). The switch can reduce or eliminate the need for skill queues or other hierarchical placement of agents or contacts, instead the switch is one of a larger population of agents within the contact center system. It can operate across the above virtual agent groups or sets of agents. Some or all aspects of the BP pairing methodology can be implemented by switches as needed, including data collection, data analysis, model generation, network flow optimization, and the like.

In some embodiments, such as those that optimize virtual agent groups, the model of agent nodes in a network flow is the contact center, regardless of whether the contact center system assigns agents to one or more skill queues. It may represent a set of agents with one or more agent skill / type combinations for agents found anywhere in the system. For example, the nodes for agents 401, 402 and 403 in FIGS. 4B-4G may represent a virtual agent group instead of individual agents, and the contacts assigned to the virtual agent group may then be individual agents within the virtual agent group. Can be assigned to (eg, random assignment, round robin assignment, model-based behavior pairing, etc.). In these embodiments, before contacts are filtered or otherwise assigned to individual skill queues and / or agent groups (eg, either contact center switch 120A or contact center switch 120B in FIG. 1). In addition, BP can be applied to contacts at a higher level within the contact center system (eg, central switch 101 in FIG. 1).

Applying BP early in the process, as long as it avoids the scripts and other normative techniques used by traditional central switches to determine which queue / switch / VDN a contact should be assigned to. Can be advantageous. These scripts and other normative techniques optimize overall contact center performance and target agent utilization of desired (eg, balanced agent utilization, minimal agent utilization imbalance, agent utilization distortion). It can be inefficient and suboptimal, both in achieving a given amount).

At this point, it should be noted that behavioral pairing in a contact center system, according to the present disclosure as described above, may involve some processing of input data and generation of output data. This input data processing and output data generation can be implemented in hardware or software. For example, certain electronic components may be employed in behavioral pairing modules, or similar or related circuits, to implement the functionality associated with behavioral pairing in a contact center system according to the present disclosure described above. Alternatively, one or more processors operating according to instructions may implement the functionality associated with behavioral pairing in the contact center system according to the present disclosure described above. If so, such instructions can be stored on one or more non-temporary processor readable storage media (eg, magnetic disks or other storage media) or on one or more carrier waves. It is within the scope of the present disclosure that it may be transmitted to one or more processors via one or more embedded signals.

The present disclosure is not limited in scope by the particular embodiments described herein. Indeed, various other embodiments of the present disclosure and modifications to the present disclosure will be apparent to those skilled in the art from the above description and accompanying drawings, in addition to those described herein. Therefore, such other embodiments and modifications are intended to fall within the scope of this disclosure. Also, the present disclosure has been described herein in the context of at least one particular implementation in at least one particular environment for at least one particular purpose, but the convenience of the present disclosure is limited thereto. Those skilled in the art will recognize that this is not done and that the disclosure can be beneficially implemented in any number of environments for any number of purposes. Therefore, the scope of claims set forth below should be construed in light of the overall breadth and intent of the present disclosure as described herein.

Claims (36)

A method for behavioral pairing in a contact center system
At least one computer processor is to determine the plurality of agents available for connection to a contact, the at least one computer processor being communicably coupled to the contact center system and being communicable. The contact center system is configured to perform improved pairing between contact agents.
Met that determine the agent pairing - the at least one computer processor, based on the type or skill of the contacts, a plurality of preferred contact from a plurality of pairing possible between the contact and the plurality of agents To determine the plurality of preferred contact-agent pairings,
(I) The at least one computer processor determines the behavioral pairing payout matrix.
(Ii) After (i), the at least one computer processor makes a targeted utilization for each agent among the plurality of agents based on the behavior pairing payout matrix, and a contact arriving at the contact center system. Determining the ratio of the number of contacts of a particular type to the contact center system to the total number of
(Iii) The plurality of computer processors, after (ii), determine a supply for each agent of the plurality of agents and a request for the type of contact. The supply for each agent of the agents is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact is for a contact arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type coming into the contact center system to the total number.
(Iv) The plurality of computer processors, after (iii), by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. Balancing the use of agents and
(V) The at least one computer processor, after (iv), said, based on the tuned supply for each agent of the plurality of agents and the tuned request for the type of contact. Determining multiple preferred contact-agent pairings
Including, and
The plurality of computer processors according to the plurality of optimal flow solutions defined by the stochastic network flow model by applying the amount of agent utilization distortion so that the at least one computer processor targets balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the preferred contact-agent pairings, the stochastic network flow model is said to follow the plurality of preferred contact-agent pairings. The optimized performance of the contact center system , generated by at least one computer processor, is due to the stochastic network flow model, each of the plurality of optimal flow solutions being the plurality of preferred contact-agent pairs. That it is associated with the corresponding one of the rings,
The at least one computer processor outputs the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. Including methods. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly than when the probabilistic network flow model is not configured to balance agent usage. The method according to claim 1, which is configured as described above. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. The method of claim 1, wherein the method is configured to optimize the overall expected value. The method of claim 3, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. The method of claim 1, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. The method of claim 1, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data. A system for behavioral pairing in a contact center system, wherein the system comprises at least one computer processor, the at least one computer processor being communicably coupled to the contact center system. The contact center system is configured to perform improved pairing between contact agents.
The at least one computer processor
Determining multiple agents available for connecting to a contact,
Determining a plurality of preferred contact-agent pairings from a plurality of possible pairings between the contact and the plurality of agents based on the type or skill of the contact , said at least one computer processor. Is
(I) Determining the behavioral pairing payout matrix and
(Ii) After (i), based on the behavior pairing payout matrix, the target utilization for each agent among the plurality of agents and the contact center system for the total number of contacts arriving at the contact center system. Determining the percentage of the number of certain types of contacts that come in,
(Iii) After (ii), determining the supply for each agent of the plurality of agents and the request for the type of contact, for each agent of the plurality of agents. The supply is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact arrives at the contact center system for the total number of contacts arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type to be made.
(Iv) (iii) is followed by balancing the use of the plurality of agents by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. When,
(V) After (iv), the plurality of preferred contact-agent pairings based on the coordinated supply for each agent of the plurality of agents and the coordinated requirements for the type of contact. To decide
It is configured to determine the plurality of preferred contact-agent pairings by performing the above .
Of the multiple preferred contact-agent pairings, according to the multiple optimal flow solutions defined by the stochastic network flow model, by applying the amount of agent utilization distortion to target balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the above, the stochastic network flow model is generated by the at least one computer processor according to the plurality of preferred contact-agent pairings. The optimized track record of the contact center system is due to the stochastic network flow model, and each of the plurality of optimal flow solutions is a corresponding one of the plurality of preferred contact-agent pairings. Is associated with, and
It is further configured to output the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. There is a system. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly than when the probabilistic network flow model is not configured to balance agent usage. 7. The system according to claim 7, which is configured as such. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. The system of claim 7, wherein the system is configured to optimize the overall expected value. The system of claim 9, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. The system of claim 7, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. The system of claim 7, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data. A product for behavioral pairing in a contact center system, said product.
Non-temporary processor readable media and
With instructions stored on the medium
The instructions are configured to be readable from the medium by at least one computer processor, the at least one computer processor being communicably coupled to the contact center system and said contact. It is configured to perform improved pairing between contact agents in the center system, thereby performing improved pairing.
Determining multiple agents available for connecting to a contact,
Determining a plurality of preferred contact-agent pairings from a plurality of possible pairings between the contact and the plurality of agents based on the type or skill of the contact , said at least one computer processor. Is
(I) Determining the behavioral pairing payout matrix and
(Ii) After (i), based on the behavior pairing payout matrix, the target utilization for each agent among the plurality of agents and the contact center system for the total number of contacts arriving at the contact center system. Determining the percentage of the number of certain types of contacts that come in,
(Iii) After (ii), determining the supply for each agent of the plurality of agents and the request for the type of contact, for each agent of the plurality of agents. The supply is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact arrives at the contact center system for the total number of contacts arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type to be made.
(Iv) (iii) is followed by balancing the use of the plurality of agents by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. When,
(V) After (iv), the plurality of preferred contact-agent pairings based on the coordinated supply for each agent of the plurality of agents and the coordinated requirements for the type of contact. To decide
To determine the plurality of preferred contact-agent pairings, and
Of the multiple preferred contact-agent pairings, according to the multiple optimal flow solutions defined by the stochastic network flow model, by applying the amount of agent utilization distortion to target balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the above, the stochastic network flow model is generated by the at least one computer processor according to the plurality of preferred contact-agent pairings. The optimized track record of the contact center system is due to the stochastic network flow model, and each of the plurality of optimal flow solutions is a corresponding one of the plurality of preferred contact-agent pairings. Is associated with, and
The at least one such as outputting the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. A manufactured product that operates a computer processor. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly compared to when the probabilistic network flow model is not configured to balance agent usage. The manufactured product according to claim 13, which is configured as described above. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. 13. The manufactured product according to claim 13, which is configured to optimize the overall expected value. The manufactured product according to claim 15, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. 13. The product of claim 13, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. 13. The product of claim 13, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data. A method for behavioral pairing in a contact center system
At least one computer processor is to determine the plurality of contacts available for connection to the agent, the at least one computer processor being communicably coupled to the contact center system and being communicable. The contact center system is configured to perform improved pairing between contact agents.
Met that determine the agent pairing - the at least one computer processor, based on the type or skill of the agent, more preferred contact from a plurality of pairing possible between said plurality of contact between the agent To determine the plurality of preferred contact-agent pairings,
(I) The at least one computer processor determines the behavioral pairing payout matrix.
(Ii) After (i), the at least one computer processor makes a targeted utilization for each agent among the plurality of agents based on the behavior pairing payout matrix, and a contact arriving at the contact center system. Determining the ratio of the number of contacts of a particular type to the contact center system to the total number of
(Iii) The plurality of computer processors, after (ii), determine a supply for each agent of the plurality of agents and a request for the type of contact. The supply for each agent of the agents is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact is for a contact arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type coming into the contact center system to the total number.
(Iv) The plurality of computer processors, after (iii), by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. Balancing the use of agents and
(V) The at least one computer processor, after (iv), said, based on the tuned supply for each agent of the plurality of agents and the tuned request for the type of contact. Determining multiple preferred contact-agent pairings
Including, and
The plurality of computer processors according to the plurality of optimal flow solutions defined by the stochastic network flow model by applying the amount of agent utilization distortion so that the at least one computer processor targets balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the preferred contact-agent pairings, the stochastic network flow model is said to follow the plurality of preferred contact-agent pairings. The optimized performance of the contact center system , generated by at least one computer processor, is due to the stochastic network flow model, each of the plurality of optimal flow solutions being the plurality of preferred contact-agent pairs. That it is associated with the corresponding one of the rings,
The at least one computer processor outputs the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. Including methods. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly than when the probabilistic network flow model is not configured to balance agent usage. 19. The method of claim 19. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. 19. The method of claim 19, which is configured to optimize the overall expected value. 21. The method of claim 21, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. 19. The method of claim 19, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. 19. The method of claim 19, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data. A system for behavioral pairing in a contact center system, wherein the system comprises at least one computer processor, the at least one computer processor being communicably coupled to the contact center system. The contact center system is configured to perform improved pairing between contact agents.
The at least one computer processor
Determining the multiple contacts available for connecting to an agent,
Determining a plurality of preferred contact-agent pairings from a plurality of possible pairings between the agent and the plurality of contacts based on the type or skill of the agent , said at least one computer processor. Is
(I) Determining the behavioral pairing payout matrix and
(Ii) After (i), based on the behavior pairing payout matrix, the target utilization for each agent among the plurality of agents and the contact center system for the total number of contacts arriving at the contact center system. Determining the percentage of the number of certain types of contacts that come in,
(Iii) After (ii), determining the supply for each agent of the plurality of agents and the request for the type of contact, for each agent of the plurality of agents. The supply is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact arrives at the contact center system for the total number of contacts arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type to be made.
(Iv) (iii) is followed by balancing the use of the plurality of agents by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. When,
(V) After (iv), the plurality of preferred contact-agent pairings based on the coordinated supply for each agent of the plurality of agents and the coordinated requirements for the type of contact. To decide
It is configured to determine the plurality of preferred contact-agent pairings by performing the above .
Of the multiple preferred contact-agent pairings, according to the multiple optimal flow solutions defined by the stochastic network flow model, by applying the amount of agent utilization distortion to target balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the above, the stochastic network flow model is generated by the at least one computer processor according to the plurality of preferred contact-agent pairings. The optimized track record of the contact center system is due to the stochastic network flow model, and each of the plurality of optimal flow solutions is a corresponding one of the plurality of preferred contact-agent pairings. Is associated with, and
It is further configured to output the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. There is a system. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly than when the probabilistic network flow model is not configured to balance agent usage. 25. The system according to claim 25. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. 25. The system of claim 25, which is configured to optimize the overall expected value. 27. The system of claim 27, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. 25. The system of claim 25, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. 25. The system of claim 25, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data. A product for behavioral pairing in a contact center system, said product.
Non-temporary processor readable media and
With instructions stored on the medium
The instructions are configured to be readable from the medium by at least one computer processor, the at least one computer processor being communicably coupled to the contact center system and said contact. It is configured to perform improved pairing between contact agents in the center system, thereby performing improved pairing.
Determining the multiple contacts available for connecting to an agent,
Determining a plurality of preferred contact-agent pairings from a plurality of possible pairings between the agent and the plurality of contacts based on the type or skill of the agent , said at least one computer processor. Is
(I) Determining the behavioral pairing payout matrix and
(Ii) After (i), based on the behavior pairing payout matrix, the target utilization for each agent among the plurality of agents and the contact center system for the total number of contacts arriving at the contact center system. Determining the percentage of the number of certain types of contacts that come in,
(Iii) After (ii), determining the supply for each agent of the plurality of agents and the request for the type of contact, for each agent of the plurality of agents. The supply is determined based on the targeted utilization for each agent of the plurality of agents, and the request for the type of contact arrives at the contact center system for the total number of contacts arriving at the contact center system. It is determined based on the ratio of the number of contacts of the particular type to be made.
(Iv) (iii) is followed by balancing the use of the plurality of agents by coordinating the supply for each agent of the plurality of agents with the requirements for the type of contact. When,
(V) After (iv), the plurality of preferred contact-agent pairings based on the coordinated supply for each agent of the plurality of agents and the coordinated requirements for the type of contact. To decide
It is configured to determine the plurality of preferred contact-agent pairings by performing the above .
Of the multiple preferred contact-agent pairings, according to the multiple optimal flow solutions defined by the stochastic network flow model, by applying the amount of agent utilization distortion to target balanced agent utilization. By optimizing the performance of the contact center system by selecting one of the above, the stochastic network flow model is generated by the at least one computer processor according to the plurality of preferred contact-agent pairings. The optimized track record of the contact center system is due to the stochastic network flow model, and each of the plurality of optimal flow solutions is a corresponding one of the plurality of preferred contact-agent pairings. Is associated with, and
The at least one such as outputting the selected one of the plurality of preferred contact-agent pairings in order to connect the corresponding contact and the corresponding agent in the contact center system. A manufactured product that operates a computer processor. The probabilistic network flow model balances agent usage so that the plurality of agents are used more evenly compared to when the probabilistic network flow model is not configured to balance agent usage. 31. The product according to claim 31, which is configured as described above. The probabilistic network flow model maximizes the overall expectation of at least one contact center metric, regardless of whether the proven agent is paired with the contact more often than the less successful agent. 31. The product of claim 31, which is configured to optimize the overall expected value. The manufactured product of claim 33, wherein the at least one contact center metric is at least one of revenue generation, customer satisfaction, and average processing time. 31. The product of claim 31, wherein the plurality of optimal flow solutions are defined based on agent skill and contact skill needs. 31. The product of claim 31, wherein the probabilistic network flow model incorporates expected payoff values based on analysis of at least one of historical contact-agent result data and contact attribute data.